Boneyard Tools

Audio Volume Normalizer

Even out the loudness of an audio file right in your browser. Drop in an MP3, WAV, M4A, or OGG and the tool measures the level, then scales the whole clip so it hits the target you pick. Use Peak mode to push the loudest sample up to just below clipping, or RMS mode to match the average energy of a track, which is closer to how loud something feels. It all runs on your device, nothing is uploaded, and the result downloads as a WAV.

How to normalize audio volume

  1. Drop in an audio file, or click to browse for one.
  2. Choose Peak or RMS mode and set the target level in dBFS.
  3. Preview the normalized clip, then download it as a WAV file.

Examples

Lift a quiet voice memo

memo.m4a peaking at about -18 dBFS
A WAV scaled so the loudest moment sits at -1 dBFS, clearly louder but not clipped.

Frequently asked questions

Is my audio uploaded anywhere?

No. Nothing is uploaded. The file is decoded, measured, and normalized entirely in your browser using the Web Audio API, so it never leaves your device. The result is encoded and downloaded locally.

What is the difference between Peak and RMS?

Peak mode scales so the single loudest sample reaches your target, which guarantees no clipping. RMS mode scales by the average energy of the clip, so two tracks normalized to the same RMS feel similarly loud. RMS can push peaks past the ceiling, so those samples are clamped.

What target level should I use?

For a safe master, peak normalize to about -1 dBFS to leave a little headroom. For matching the perceived loudness of several clips, RMS normalize them all to the same value, commonly around -14 to -12 dBFS. The default is -1 dBFS.

What audio formats are supported?

MP3, WAV, M4A, FLAC, OGG, and other common formats all work, because the browser decodes them before processing. You can also drop in the audio track of an MP4 or WebM. The output is always a WAV file.

Will normalizing a silent file do anything?

No. You cannot scale silence up to a target, so a fully silent file is returned unchanged. Normalization only helps when there is real signal to measure and boost or attenuate.

Related tools